Method and device for detecting wheels with deformed treads in railroad vehicles
Device for detecting defective wheels on rail cars
Apparatus for the acoustic detection of defects in a moving strip
Wheel inspection system Patent #: 6523411
ApplicationNo. 10772535 filed on 02/06/2004
US Classes:246/169S, Sound detectors246/169R, Train defect324/718, To detect a flaw or defect73/146, TIRE, TREAD OR ROADWAY73/639, Transducer forms wheel or is within a wheel73/620By reflected wave
ExaminersPrimary: Le, Mark T.
Attorney, Agent or Firm
Foreign Patent References
International ClassB61K 9/00
The present invention relates to a system for detection of railroad and wheel defectswherein rail defect monitors are installed in a segment of the rails in acoustically isolated relationship provide information identifying wheels with defects, if desired, from the remote locations of the monitors to a central monitoring station.
BACKGROUND OF THE INVENTION
Railroads are important transportation systems for moving passengers and freight. The railroad car wheels are subject to wear and to damage due to material defects, and even sabotage. Many defects worsen gradually and, if detected early, can berepaired or replaced before accidents occur.
An automatic system for detection of such defects, aside of its value as a means for preventing accidents, injuries and deaths, would also allow for train maintenance schedule to be done as indicated by the condition of the wheels, rather thanjust based on past experience or regulations--an opportunity for cost-saving.
The advent of inexpensive and reliable sensors, microprocessors and electronics makes such automated systems practical and cost-effective.
SUMMARY OF THE INVENTION
The invention described here for automated detection of railroad car wheel defects has advantages over prior art in that it is not affected by electro-magnetic interference, can be conveniently retrofitted to the existing rails, and will senseand report a wide range of different defects and the location of the wheels so affected.
The system uses analysis of the acoustical spectrum of sounds generated by the rolling train wheels. The harmonics content in such spectrum changes as a defective wheel rolls over a rail segment equipped with and acoustic or vibration sensors. The intensity distribution of the harmonics is related to the nature and the extent of a defect, hence spectral analysis will reveal the nature and the severity of a defect.
In the preferred embodiment of this invention one or more acoustically isolated rail segments are installed into both rails. Each such segment is equipped with an acoustical/vibration transducer capable of picking up any sounds or vibrations aswheel rolls over this segment or segments. The length of each such rail segment is such that it can monitor a wheel through about one half of its full rotation and the following rail segment would monitor the second half. In this manner only one wheelcan be rolling over the rail segments at one time, preventing signal interference.
A defective wheel generates sounds or causes vibrations that are distinctly different from sounds or vibrations generated by wheels without defects. Since one can expect the sounds or vibrations to be cyclic, the difference is in the intensitydistribution of the harmonic content that can be conveniently analyzed using Fast Fourier Transform to obtain the frequency distribution of these harmonics and then to compare it reference signal patterns stored in the detection system. If thecomparison indicates that the defect or wear is severe enough for the wheel to be replaced, and appropriate signal indicating the position of the wheel in the train and, possibly, the nature of the defect is transmitted to a central location.
Alternately, the detection system could be equipped with a paint sprayer that would upon the appropriate system spray the suspected wheel to mark it for further inspection and, if necessary, replacement.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic diagram of the stationary defect detection system to be installed in the rails.
FIGS. 2A and 2B are top views of the two of the alternatives for the acoustical gap in the rails, which would minimize the wear of the rails
FIG. 3 is a side view of the rail segments used for detection of wheel defects
FIG. 4 is a schematic diagram of one version of the power supply
FIG. 5 depicts an alternative installation of sensors in the rail sections for sensing
FIG. 6 shows in perspective a fragment of the rail section to be inserted in FIG. 5
FIG. 7 is perspective view of a segment of a channel structure used to join the sensing rail section of FIG. 5 with the rest of the track.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring now to FIG. 1, which is a schematic diagram of the wheel defect detection system, 102 and 103 are the rail segments. The train wheels are 101, 104 and 105 are the acoustical/vibration sensors conveniently attached to the rail segments102 and 103, respectively. These sensors pick the sounds and vibrations generated when the train wheels rolls over the rails. The intensity vs. frequency distribution of these sounds/vibrations is defined by the speed of the train the condition of therail and wheels surfaces and of the joints between the rails. Any anomalies in the rails produce a significant change in the sound/vibration spectrum, such as when a wheel is out of round. The sounds and vibrations propagate through the rail segments102 and 103 to one of the sensors 104 or 105, which also include amplifiers.
The electrical signals corresponding to sounds/vibrations picked up by sensors 104 and/or 105 are amplified and fed into module 106 which contains an integrated acoustical spectrum analyzer that uses the Fast Fourier Transform algorithm toperform the conversion of the signals from time domain into frequency domain and thus generate an intensity vs. frequency distribution. Since more than a pair of sensors may be used along the rail line to detect wheel defect, each sensor also transmitsan I.D. code to identify its location.
The information processed in the Fast Fourier Transform section of the module 106 is fed into a microprocessor in the same module, which compares the received spectrum data to a stored reference spectrum representing defect-free wheels andisolates the changes. These changes are then used in the microprocessor to determine the nature of the defect and its extent with reference to stored data on all typical defects and anomalies. The resulting diagnosis data are then sent via transmitter107 to a central monitoring station. Alternately, the data could be transmitted over a wire. To identify the defective wheel, the system is equipped with a counter that counts the number of wheels passing over the rail segments from the first wheel inthe locomotive to the defective wheel or wheels in the train.
Instead of sending information to the central monitoring station, the signal indicating the presence of a defective wheel can also be used to trigger a paint sprayer that would mark the defective wheel for further inspection and, if need be,replacement at a repair facility.
FIGS. 2A and 2B are the top views of the acoustic gaps associated with the rail segments that are equipped with sound/vibration sensors. The shape of the gaps is designed to reduce rail wear at these points. It is to be understood that othershapes of acoustic gaps intended to achieve the same effect are within the scope of this invention.
FIG. 3 illustrates the side view of the rail segments 102 and 103 and the acoustic gaps 108, 109, and 110. Flanges 201, 202, and 203 are used to join the rail segments into the rail line. To minimize the transmission of the sound/vibrationsbetween the rail segments and the rail line, the fasteners 204, have associated with them cylindrical inserts made of a hard material, such as a fiber-reinforced polymer, which is a poor conductor of higher frequency sounds/vibration associated withdefective wheels.
The system depicted in FIG. 1 may be self-powered or powered by a long-lasting battery The power is generated as follows: Whenever a train travels over rails where sensors 104 and 105 are located, each passing wheel generates vibrations andsound, which are transducted by the sensors into electrical signals, amplified and fed into a pair of isolating transformers 301 and 302 as shown in FIG. 4. These transformers are preferably high-frequency units using toroidal cores. The signals arerectified by diodes 303 and 304. The resulting direct current charges the battery 305, which provides the power for the system, where the power is provided by a primary, long lasting battery, such as a lithium battery. The battery is periodicallyreplaced or recharged.
FIGS. 5, 6 and 7 illustrate an alternative arrangement for inserting a rail section equipped with a vibration/acoustic sensor into the rail track. A section of a channel 502 is used to join the rails. The channel 502 has on the bottom a layerof sound/vibration isolating material 503 and the sensor 504. It should be understood that the method of joining depicted in FIGS. 5, 6 and 7 may be combined with rail configurations of FIGS. 2A and 2B to minimize the wear of the rails.
Since the defect detection system is stationary installed at a known location, it can be uniquely identified when transmitting a warning signal by an appropriate fixed code.
* * * * *